root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-03 15:24:35 +00:00
Code Issues Actions 7 Packages Projects Releases Wiki Activity

Commit Graph

  • 5d4158b12c
         Add testing to matrix Henri Vasserman 2023-05-02 16:54:28 +0300
  • 2d099e5193
         ggml: add names to tensors (#1268) master-2d099e5 slaren 2023-05-02 16:03:00 +0200
  • 069b3d4c37 Adds --mlock argument Sergey Kucher 2023-05-02 16:19:37 +0300
  • 0afc2f91db check for thread support in quantize-stats John Doe 2023-05-02 08:29:13 -0400
  • a0de04a6f1
         Add OpenBLAS support Henri Vasserman 2023-05-02 13:21:54 +0300
  • a48eebe621
         Add OpenCL and CLBlast support Henri Vasserman 2023-05-02 00:19:10 +0300
  • 5a10ea50da up ver Concedo 2023-05-02 18:19:08 +0800
  • 9a9b217e57 updated embedded kobold lite with multiuser chat Concedo 2023-05-02 18:18:05 +0800
  • b8279c82d0 When loading a safetensors file, ignore the metadata header; If no pt or pth files are available, attempt to load safetensors files ubik2 2023-05-02 02:47:35 -0700
  • 308550b789 Adding the ability to have zero '0' as a seed number. RB 2023-05-02 04:27:30 -0400
  • 6f702f2700 fixed stop sequence crash Concedo 2023-05-02 14:56:50 +0800
  • 94827172e0 Merge branch 'master' into concedo Concedo 2023-05-02 14:38:31 +0800
  • 433fa1e8b2 fix for stop sequence missing, added print for exception when loading GUI Concedo 2023-05-02 14:18:04 +0800
  • bcc7f8b6bf Merge remote-tracking branch 'upstream/master' into eval-thread-count ml6 2023-05-01 22:05:43 -0700
  • 14fa3d108b Handle C++ libraries without threading support. John Doe 2023-05-01 06:52:01 -0400
  • 9af4b9e43f minor mprovements to dot file formatting Slaren 2023-05-01 23:39:53 +0200
  • bc1c13bb66
         train with two examples, creating new tensors each time.. xaedes 2023-05-01 22:22:00 +0200
  • fcbc262eb9
         Merge 'origin/master' into hipblas Henri Vasserman 2023-05-01 22:45:29 +0300
  • 5f23052eb2
         switching from training with adam to lbfgs produces much better results in the baby-llama example xaedes 2023-05-01 21:01:17 +0200
  • 29a0f8b940
         fix softmax in baby-llama example xaedes 2023-05-01 20:02:48 +0200
  • 8fde656d24
         add baby-llama example training a very small llama model from scratch to output a sinusoidal wave. xaedes 2023-05-01 19:30:04 +0200
  • 0703cdf2eb remove cloudflare insights Concedo 2023-05-02 00:38:10 +0800
  • f4cef87edf
         Add git-based build information for better issue tracking (#1232) master-f4cef87 DannyDaemonic 2023-05-01 09:23:47 -0700
  • 52151b6dac ggml: add names to tensors Slaren 2023-05-01 17:17:03 +0200
  • 58b367c2d7
         cuBLAS: refactor and optimize f16 mat mul performance (#1259) master-58b367c slaren 2023-05-01 18:11:07 +0200
  • ea3a0ad6b6
         llama : update stubs for systems without mmap and mlock (#1266) master-ea3a0ad xloem 2023-05-01 08:58:51 -0400
  • a79756b210 cuBLAS: update block_q5_1 Slaren 2023-05-01 14:51:03 +0200
  • 1c4dc1e498
         update quantization types in switch-case of add_at and add1 xaedes 2023-05-01 14:30:29 +0200
  • 72bcfb50c8
         successfully test backward pass of repeat xaedes 2023-05-01 01:11:41 +0200
  • 8b5b2f089e
         fix backward pass for repeat xaedes 2023-05-01 01:11:12 +0200
  • ba62c79bd5
         add missing GGML_OP_SUM_ROWS xaedes 2023-05-01 14:29:52 +0200
  • c4539ede53
         add operation ggml_sum_rows xaedes 2023-05-01 01:10:30 +0200
  • 2277053839
         add todos for llama backward pass xaedes 2023-04-30 21:42:52 +0200
  • 2ecc690980
         successfully test backward pass of rms_norm xaedes 2023-04-30 21:39:03 +0200
  • 84a4b39917
         fix backward pass for rms_norm xaedes 2023-04-30 21:34:21 +0200
  • b18b72da00
         successfully test backward pass of view_1d, view_2d and view_3d xaedes 2023-04-30 17:22:25 +0200
  • 84436383eb
         fix view backward pass xaedes 2023-04-30 17:21:21 +0200
  • f0302fa71b
         successfully test get_rows backward xaedes 2023-04-28 20:32:00 +0200
  • 96e773bbde
         fix get rows backward pass xaedes 2023-04-28 20:31:36 +0200
  • 7281f60572
         move dup call into the actual add_at functions xaedes 2023-04-28 20:30:42 +0200
  • 3dbd649cf9
         fix diag_mask to work with non-inplace input xaedes 2023-04-28 20:03:56 +0200
  • b9920e5c3e
         test-grad0 : fix test for div xaedes 2023-04-28 20:00:25 +0200
  • 19f51592b5
         successfully test diag_mask_inf and diag_mask_zero backward xaedes 2023-04-28 18:43:58 +0200
  • d42531fa56
         fix comments xaedes 2023-04-28 18:22:40 +0200
  • 1997152f7f
         test-grad0.c add TODO for view_2d and view_3d xaedes 2023-04-28 18:16:55 +0200
  • c601df973c
         successfully test transpose backward and permute for all permutations xaedes 2023-04-28 18:14:37 +0200
  • 3d21f2646e
         implement ggml_cont backward pass xaedes 2023-04-28 18:12:25 +0200
  • 02d3fd0894
         fix sub, mul and div functions to work correctly with transposed tensors xaedes 2023-04-28 18:11:26 +0200
  • b0555fce95
         some minor test-grad0 fixes xaedes 2023-04-28 17:47:53 +0200
  • a7a837047c
         successfully test permute backward xaedes 2023-04-28 17:47:23 +0200
  • 86b44a02e4
         test-grad0.c : add print_elements to help with debugging xaedes 2023-04-28 17:46:55 +0200
  • 339b2adf48
         fix ggml_forward_add1 functions to work correctly with transposed tensors xaedes 2023-04-28 17:43:50 +0200
  • b9416d71f8
         fix ggml_forward_add functions to work correctly with transposed tensors xaedes 2023-04-28 17:42:24 +0200
  • 410a47a79e
         minor code format improvement xaedes 2023-04-27 17:00:40 +0200
  • 124fdca973
         successfully test view backward xaedes 2023-04-28 18:36:07 +0200
  • cecd6c7665
         bug fix for add_at forward xaedes 2023-04-27 16:58:22 +0200
  • 83fa6b3bcb
         fix ggml_compute_forward_dup_same_cont for when nelements < nthreads xaedes 2023-05-01 14:42:44 +0200
  • 2bdc09646d
         ggml : fix ggml_used_mem() (#1264) master-2bdc096 Kerfuffle 2023-05-01 05:56:07 -0600
  • 70269cae37
         llama : fix session load / save (#1263) master-70269ca Georgi Gerganov 2023-05-01 14:54:59 +0300
  • 3e023c59b8 update stubs for systems without mmap and mlock John Doe 2023-05-01 07:45:54 -0400
  • 4cd0a480bf fix build Slaren 2023-05-01 00:12:46 +0200
  • a9ad140c17 cuBLAS: use multiple streams, choose smartly between mul_mat_q and mul_mat_f16 Slaren 2023-04-30 23:46:19 +0200
  • cf93fdcfda cuBLAS: refactor, convert fp16 to fp32 on device Slaren 2023-04-30 20:38:12 +0200
  • b925f1f1b0
         cuBLAS: fall back to pageable memory if pinned alloc fails (#1233) master-b925f1f slaren 2023-05-01 13:32:22 +0200
  • acae41c7ac ggml_used_mem can segfault if called before any objects are created. KerfuffleV2 2023-05-01 04:21:33 -0600
  • c0335b51f9
         llama : fix session load / save Georgi Gerganov 2023-05-01 11:08:09 +0300
  • 90b19bd6ee
         llama : let context be const when accessing const data (#1261) master-90b19bd Alex Klinkhamer 2023-05-01 00:24:20 -0700
  • 4d38795563 add UI for token unbanning Concedo 2023-05-01 12:10:21 +0800
  • 3de34ee492 Merge branch 'master' into concedo_experimental Concedo 2023-05-01 12:03:46 +0800
  • 560dacedbd update readme Concedo 2023-05-01 11:41:25 +0800
  • 52892512cd Short hash, less fancy Makefile, and don't modify build-info.h if it wouldn't change it Danny Daemonic 2023-04-30 18:39:33 -0700
  • efe84ca371
         llama : let context be const when accessing const data grencez 2023-04-30 19:50:17 -0700
  • ddb3e88376 4 space indenting for cmake, attempt to clean up my mess in Makefile Danny Daemonic 2023-04-30 00:48:23 -0700
  • 3e7ec6e1e2 Broke out build-info.cmake, added find_package fallback, and added build into to all examples, added dependencies to Makefile Danny Daemonic 2023-04-29 15:20:21 -0700
  • d2b6d2ce39 Fix conditional dependency on missing target Danny Daemonic 2023-04-29 12:13:57 -0700
  • eac5d689dc Redo "CMAKE_CURRENT_SOURCE_DIR" and clearer build messages Danny Daemonic 2023-04-29 09:57:22 -0700
  • db0b8357b6 "build (hash)" and "CMAKE_SOURCE_DIR" changes Danny Daemonic 2023-04-29 09:09:51 -0700
  • fbcfc8446c macOS fix Danny Daemonic 2023-04-29 06:28:39 -0700
  • 8272f4aed8 Add git-based build information for better issue tracking Danny Daemonic 2023-04-29 05:55:22 -0700
  • c1a8893de3
         de-duplicate ggml_forward_dup code taking care of contiguous tensors of same type. xaedes 2023-04-27 16:55:22 +0200
  • 38675e537c
         add shape annotations for llama xaedes 2023-04-27 16:39:41 +0200
  • 93106504fd
         align shape annotations xaedes 2023-04-27 00:21:31 +0200
  • fea42be47a
         successfully test soft_max backward xaedes 2023-04-27 00:16:18 +0200
  • 1a80e9a0fa
         correctly implement softmax backward pass using new operation ggml_diag xaedes 2023-04-27 00:13:43 +0200
  • 54ab300cc4
         add test-opt.c xaedes 2023-04-26 21:35:20 +0200
  • ecf949b10f
         successfully test reshape backward xaedes 2023-04-26 20:34:33 +0200
  • c483a7dac5
         bug fix for reshape backward pass xaedes 2023-04-26 20:34:08 +0200
  • b2bd8222da
         successfully test cpy backward xaedes 2023-04-26 20:14:52 +0200
  • 0ea8201c86
         bug fix for cpy backward pass xaedes 2023-04-26 20:14:33 +0200
  • 7571147242
         successfully test rope backward xaedes 2023-04-26 00:46:49 +0200
  • b583136cfa
         improve performance of sqr backward pass xaedes 2023-04-26 00:46:20 +0200
  • bfe507213c
         improve performance of sum backward pass xaedes 2023-04-26 00:43:02 +0200
  • 0197bcb0ff
         successfully test scale backward xaedes 2023-04-25 22:26:26 +0200
  • a367eb9eda
         bug fix for scale backward pass xaedes 2023-04-25 22:25:53 +0200
  • 671e5922e2
         successfully test silu backward xaedes 2023-04-25 22:06:05 +0200
  • 6fb08b4554
         bug fixes for silu_back xaedes 2023-04-25 22:05:45 +0200
  • 9d6fc28f18
         disable graph dot export as it floods console xaedes 2023-04-25 22:05:22 +0200
  • 9345f4c3a5
         test both gradients of mul_mat xaedes 2023-04-24 22:37:09 +0200
  • 20e3c1d2b4
         use GGML_PRINT_DEBUG for debug messages which will otherwise flood the console xaedes 2023-04-24 21:21:50 +0200
  • 0da26753fd
         add test-grad0.c xaedes 2023-04-25 21:32:05 +0200